Linux-mm Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Christian Brauner" <brauner@kernel.org>, "Jan Kara" <jack@suse.com>
Cc: "Al Viro" <viro@zeniv.linux.org.uk>,
	linux-ext4@vger.kernel.org, linux-mm@kvack.org,
	linux-efi@vger.kernel.org, linux-mtd@lists.infradead.org,
	dsterba@suse.cz, tytso@mit.edu, "Mark Fasheh" <mark@fasheh.com>,
	"Joel Becker" <jlbec@evilplan.org>,
	"Joseph Qi" <joseph.qi@linux.alibaba.com>
Subject: Re: [PATCH v2 02/34] ecryptfs: fall back to a read-only lower file
Date: Fri, 30 Oct 2026 16:09:48 +0000	[thread overview]
Message-ID: <sashiko-outbox-191221@kernel.org> (raw)
In-Reply-To: <20261029-work-fs-sb_rdonly-v2-2-2f44f64f0b66@kernel.org>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] ecryptfs: ECRYPTFS_KEY_VALID left set after ecryptfs_initialize_file() fails
--

commit f7dbc785a33f0a188b62f1d0ae6c24e5fc32a62c
Author: Christian Brauner <brauner@kernel.org>

ecryptfs: fall back to a read-only lower file

This commit falls back to opening the lower file read-only when opening it
read-write fails with EROFS, and rejects writes to a read-only lower file in
ecryptfs_write_lower() and __ecryptfs_truncate() with EROFS.

> diff --git a/fs/ecryptfs/read_write.c b/fs/ecryptfs/read_write.c
> index b3b451c2b9410..b3956ae50ff66 100644
> --- a/fs/ecryptfs/read_write.c
> +++ b/fs/ecryptfs/read_write.c
> @@ -33,6 +33,9 @@ int ecryptfs_write_lower(struct inode *ecryptfs_inode, char *data,
>  	lower_file = ecryptfs_inode_to_private(ecryptfs_inode)->lower_file;
>  	if (!lower_file)
>  		return -EIO;
> +	/* ecryptfs_open_lower() may have fallen back to a read-only file. */
> +	if (!(lower_file->f_mode & FMODE_WRITE))
> +		return -EROFS;

[Severity: Medium]
When opening an existing 0-byte lower file with O_RDONLY on a read-only lower
mount (or a lower filesystem in an error state) with default mount options,
ecryptfs_open() sets ECRYPTFS_POLICY_APPLIED | ECRYPTFS_ENCRYPTED on
crypt_stat->flags and calls read_or_initialize_metadata().

Because the lower file is empty, ecryptfs_read_metadata() fails and
read_or_initialize_metadata() calls ecryptfs_initialize_file():

fs/ecryptfs/inode.c:ecryptfs_initialize_file() {
	rc = ecryptfs_new_file_context(ecryptfs_inode);
	if (rc) {
		...
		goto out;
	}
	rc = ecryptfs_get_lower_file(ecryptfs_dentry, ecryptfs_inode);
	...
	rc = ecryptfs_write_metadata(ecryptfs_dentry, ecryptfs_inode);
	if (rc)
		printk(KERN_ERR "Error writing headers; rc = [%d]\n", rc);
	ecryptfs_put_lower_file(ecryptfs_inode);
out:
	return rc;
}

ecryptfs_new_file_context() sets ECRYPTFS_ENCRYPTED | ECRYPTFS_KEY_VALID in
crypt_stat->flags before ecryptfs_write_metadata() calls
ecryptfs_write_lower() and fails with -EROFS. Neither
ecryptfs_initialize_file() nor read_or_initialize_metadata() clears
ECRYPTFS_KEY_VALID (or ECRYPTFS_POLICY_APPLIED) on failure.

On a second open(O_RDONLY) or ecryptfs_setattr() on the same cached inode,
read_or_initialize_metadata() checks the flags at the start of the function:

fs/ecryptfs/file.c:read_or_initialize_metadata() {
	if (crypt_stat->flags & ECRYPTFS_POLICY_APPLIED &&
	    crypt_stat->flags & ECRYPTFS_KEY_VALID) {
		rc = 0;
		goto out;
	}
	...
}

Because both flags remain set from the failed initialization, does the second
open short-circuit metadata validation and succeed on the uninitialized empty
file (and if the lower mount is later remounted read-write, allow writes using
an in-memory key whose header was never written to disk)?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261029-work-fs-sb_rdonly-v2-0-2f44f64f0b66@kernel.org?part=2


  reply	other threads:[~2026-10-30 16:09 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-29 16:06 [PATCH v2 00/34] fs: don't set SB_RDONLY after filesystem errors Christian Brauner
2026-10-29 16:06 ` [PATCH v2 01/34] ecryptfs: don't leak the inode when a lookup can't open the lower file Christian Brauner
2026-10-29 16:06 ` [PATCH v2 02/34] ecryptfs: fall back to a read-only " Christian Brauner
2026-10-30 16:09   ` sashiko-bot [this message]
2026-10-29 16:06 ` [PATCH v2 03/34] fs: add a generic emergency read-only state Christian Brauner
2026-10-29 16:06 ` [PATCH v2 04/34] ext2: don't set SB_RDONLY after filesystem errors Christian Brauner
2026-10-29 16:06 ` [PATCH v2 05/34] jfs: " Christian Brauner
2026-10-30 16:09   ` sashiko-bot
2026-10-29 16:06 ` [PATCH v2 06/34] nilfs2: " Christian Brauner
2026-10-30 16:09   ` sashiko-bot
2026-10-29 16:06 ` [PATCH v2 07/34] fat: " Christian Brauner
2026-10-30 16:09   ` sashiko-bot
2026-10-29 16:06 ` [PATCH v2 08/34] exfat: keep the current mount options on remount Christian Brauner
2026-10-29 16:07 ` [PATCH v2 09/34] exfat: don't set SB_RDONLY after filesystem errors Christian Brauner
2026-10-30 16:09   ` sashiko-bot
2026-10-29 16:07 ` [PATCH v2 10/34] exfat: use the generic emergency read-only state Christian Brauner
2026-10-29 16:07 ` [PATCH v2 11/34] hfs: don't set SB_RDONLY after MDB write errors Christian Brauner
2026-10-30 16:09   ` sashiko-bot
2026-10-29 16:07 ` [PATCH v2 12/34] hfs: use the generic emergency read-only state Christian Brauner
2026-10-29 16:07 ` [PATCH v2 13/34] hfsplus: don't set SB_RDONLY when creating the attributes tree Christian Brauner
2026-10-29 16:07 ` [PATCH v2 14/34] affs: don't set SB_RDONLY after filesystem errors Christian Brauner
2026-10-29 16:07 ` [PATCH v2 15/34] ufs: " Christian Brauner
2026-10-29 16:07 ` [PATCH v2 16/34] hpfs: " Christian Brauner
2026-10-30 16:10   ` sashiko-bot
2026-10-29 16:07 ` [PATCH v2 17/34] ntfs: " Christian Brauner
2026-10-30 16:10   ` sashiko-bot
2026-10-29 16:07 ` [PATCH v2 18/34] ntfs: use the generic emergency read-only state Christian Brauner
2026-10-29 16:07 ` [PATCH v2 19/34] ocfs2: keep the current mount options on remount Christian Brauner
2026-10-30  2:58   ` Joseph Qi
2026-10-29 16:07 ` [PATCH v2 20/34] ocfs2: don't set SB_RDONLY after filesystem errors Christian Brauner
2026-10-30  3:01   ` Joseph Qi
2026-10-29 16:07 ` [PATCH v2 21/34] zonefs: don't set SB_RDONLY after I/O errors Christian Brauner
2026-10-29 16:07 ` [PATCH v2 22/34] zonefs: use the generic emergency read-only state Christian Brauner
2026-10-29 16:07 ` [PATCH v2 23/34] btrfs: check BTRFS_FS_STATE_RO instead of SB_RDONLY at runtime Christian Brauner
2026-10-29 16:07 ` [PATCH v2 24/34] btrfs: don't set SB_RDONLY after filesystem errors Christian Brauner
2026-10-29 16:07 ` [PATCH v2 25/34] gfs2: wait for a pending withdraw before remounting read-only Christian Brauner
2026-10-30 16:10   ` sashiko-bot
2026-10-29 16:07 ` [PATCH v2 26/34] gfs2: don't set SB_RDONLY on withdraw Christian Brauner
2026-10-30 16:10   ` sashiko-bot
2026-10-29 16:07 ` [PATCH v2 27/34] efivarfs: stop using SB_RDONLY for read-only efivar ops Christian Brauner
2026-10-29 16:07 ` [PATCH v2 28/34] ubifs: don't set SB_RDONLY after errors Christian Brauner
2026-10-30  3:42   ` Zhihao Cheng
2026-10-30 16:10   ` sashiko-bot
2026-10-29 16:07 ` [PATCH v2 29/34] selftests/filesystems: test filesystem errors while frozen Christian Brauner
2026-10-29 16:07 ` [PATCH v2 30/34] ext4: allow remounting read-only in the emergency state Christian Brauner
2026-10-29 16:07 ` [PATCH v2 31/34] ext4: use the generic emergency read-only state Christian Brauner
2026-10-29 16:07 ` [PATCH v2 32/34] selftests/filesystems: test ext4 remounts in the emergency state Christian Brauner
2026-10-29 16:07 ` [PATCH v2 33/34] xfs: allow remounting read-only after a shutdown Christian Brauner
2026-10-29 16:07 ` [PATCH v2 34/34] selftests/filesystems: test xfs remounts " Christian Brauner

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=sashiko-outbox-191221@kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=brauner@kernel.org \
    --cc=dsterba@suse.cz \
    --cc=jack@suse.com \
    --cc=jlbec@evilplan.org \
    --cc=joseph.qi@linux.alibaba.com \
    --cc=linux-efi@vger.kernel.org \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-mtd@lists.infradead.org \
    --cc=mark@fasheh.com \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=tytso@mit.edu \
    --cc=viro@zeniv.linux.org.uk \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox